TGFBI, induced by TGF-b1, is a fascilin-like protein and has been shown to be a component of extracellular matrix in a variety of human tissues. Using in vitro transformation systems, the applicant and his co-investigators have shown previously that loss of TGFBI expression is causally linked to tumorigenic phenotype in radiation/asbestos fibers-treated human bronchial epithelial cells, which is further substantiated by the studies that reconstituted expression of TGFBI in human lung cancer cells completely abrogate their tumorigenicity. Significant downregulation of TGFBI gene was found in various human tumor cell lines and in one-third of primary human lung carcinomas. Furthermore, hypermethylation of CpG island of TGFBI promoter correlates with the gene silencing. These findings suggest a significant role of TGFBI gene in human tumor progression. Using TGFBI-/- mouse model, the applicant and his co- investigators have provided the preliminary data showing an increased predisposition to spontaneous tumors in TGFBI-/- mice as well as an enhanced cell proliferation and aberrant activation of CREB and cyclin D1 in TGFBI-/- MEFs. This raises the following questions: Is increased frequency of spontaneous tumors in TGFBI-/- mice significantly higher than in wild type and heterozygous mice? What is the potential role of cyclin D1 upregulation in cell proliferation and spontaneous tumor development in mice with disruption of TGFBI gene? And, which signaling pathway is responsible for TGFBI-regulated cyclin D1 expression? To address these issues, a series of 3 specific aims are proposed to address the 3 testable hypotheses. The proposed studies are likely to have a significant impact on our current understanding of functional role of TGFBI loss in cell proliferation and tumor progression. PROJECT NARRATIVE Interaction between integrin and extracellular matrix (ECM) plays a crucial role in the regulation of cell adhesion and proliferation, anti-apoptotic death and tumor progression. TGFBI encodes a secreted protein that contains an integrin-binding site and promotes cellular adhesion and spreading. We have shown previously that TGFBI is ubiquitously expressed in normal human tissues, whereas its expression was either decreased or lost in a variety of human tumor cell lines as well as in one-third of primary human lung tumor samples. Recovered expression of TGFBI in human cancer cells that lack endogenous TGFBI protein significantly suppresses their tumorigenicity. We further demonstrate that promoter hypermethylation, one of the mechanisms by which tumor suppressor genes are inactivated in human cancers, correlates with silencing of TGFBI Promoter. These data suggest a tumor suppressor function of TGFBI in vivo. To study this hypothesis, we have succeeded in generating TGFBI-deficient mice. Preliminary data based on this model showed that lack of TGFBI expression results in an increased tendency to spontaneous tumor development in mice, and an enhanced cell proliferation and aberrant activation of CREB and cyclin D1 in TGFBI-null cells. These raise the following questions: Is increased frequency of spontaneous tumors in TGFBI-/- mice significantly higher than in wild type and heterozygous mice? What is the potential role of cyclin D1 upregulation in an increased cell proliferation and an enhanced spontaneous tumor development in TGFBI-/- mice? And, which signaling pathway is responsible for TGFBI-regulated cyclin D1 expression? To address these issues, a series of 3 specific aims are proposed to address the 3 testable hypotheses. Hypothesis 1: Disruption of TGFBI may contribute to tumorigenic progression. Specific Aim 1A-C will address this hypothesis and use the TGFBI-/-, heterozygous and wild type mice to compare the difference in frequencies of both spontaneous tumor growth and DMBA-induced skin tumors. Hypothesis 2: The cell cycle regulator, cyclin D1, may be critical in mediating TGFBI-regulated cell proliferation and tumor progression. Aim 2A will address whether silence of TGFBI in wild type MEFs can activate cyclin D1 and induce aberrant cell proliferation and cell transformation. In Specific Aim 2B, essential role of cyclin D1 upregulation in TGFBI-/- cells will be examined by silence of cyclin D1 expression. Suppression of cell proliferation and transformation phenotypes will be defined in cyclin D1-silenced TGFBI-/- cells. And Aim 2C will further address the importance of cyclin D1 upregulation using an in vivo mouse model. Hypothesis 3: TGFBI, a secreted protein, may regulate downstream targets through an integrin-associated signaling pathway. In Aim 3A, causal role of CREB activation in cyclin D1 upregulation will be defined. Aim 3B will determine whether an aberrant activation of CREB and cyclin D1 are due to a dysregulated PKA (Protein kinase A) activity. In Aim 3C, integrin receptor(s) and functional domain of TGFBI that are involved in TGFBI- integrin interaction and in the regulation of CREB and cyclin D1 activation will be identified. And in Aim 3D, aberrant signaling found in TGFBI-/- cells will be examined in spontaneous tumor tissues arising from TGFBI-/- mice to determine their significance in process of tumor progression. The proposed studies will provide the in vivo evidence for supporting the hypothesis that TGFBI possesses anti-tumor function. Since TGFBI is frequently down-regulated in primary human tumors, the data generated from this application are likely to have a significant impact on our current understanding of functional role of TGFBI loss in development of human cancers.